Vision based Target Interception using Aerial Manipulation
This addresses a specific domain problem for UAV robotics in selective object interception, but it is incremental as it adapts existing methods to a new challenge scenario.
The paper tackles the problem of autonomously intercepting balloons at random locations using a UAV with vision and manipulation, as part of a robotic challenge, and implements the framework on hardware using ROS.
Selective interception of objects in unknown environment autonomously by UAVs is an interesting problem. In this work, vision based interception is carried out. This problem is a part of challenge 1 of Mohammed Bin Zayed International Robotic Challenge, 2020, where, balloons are kept at five random locations for the UAVs to autonomously explore, detect, approach and intercept. The problem requires a different formulation to execute compared to the normal interception problems in literature. This work details the different aspect of this problem from vision to manipulator design. The frame work is implemented on hardware using Robot Operating System (ROS) communication architecture.